In 1981, in response to a pile of letters asking him what he thought about a new baseball offense model created by the sports journalist Thomas Boswell, the father of sabermetrics, Bill James, wrote that “the world needs another offensive rating system like Custer needed more Indians (or, for that matter, like the Indians needed another Custer)…. What we really need is for the amateurs to clear the floor.”

Although cricket boffins may not understand what custard has to do with baseball, they can surely still appreciate James’ obvious disdain for offensive rating systems. That is, statistics which are designed to provide a better insight into a batsman’s ability.

Like baseball, cricket has a numerous amount of metrics which can be used to evaluate a batsman’s performances. The most basic runs tally is one of the most often cited statistics for doing this. Late in the year, during the Australian summer of cricket period between November and December, commentators will often refer to the number of runs a batsman scores “in a calendar year”.

Although this is a great opportunity to see how well any given batsman has done in the last year, the metric fails to take into account the number of matches a batsman played at home versus away. Batsmen more comfortable in their home conditions will of course score more than in unfamiliar territory.

Batting average and batting strike rate are useful statistics in longer formats of the game such as Test cricket and One-Day cricket, but their younger brother Twenty20 has already outgrown them.

Batting average should no longer be the preferred statistic for displaying a batsman’s proficiency in the shortest format of the game. This is largely thanks to the fact that the definition of the batting average and the equation used to calculate it do not match. The actual batting average equation offers a “reward” for batsmen who survive their innings and remain not out.

This means it is possible for two batsmen with significantly different performance records to record the same batting average. This leads to both players being assessed as having a similar level of performance and ability. A batsman that scores 19, 20, 25, 15 and 21 from five innings will actually have the same batting average as a batsman that scores 95, 2, 3, 0 and 0 across five innings.

Batting average is used as the primary assessor of a batsman’s ability. That’s why it’s so astounding that the batting average remains such a crude predictor of the expected run output of a batsman during a game.

Now, we are not suggesting that the batting average should not be removed entirely. It can still provide valuable information. For example, Brisbane Heat’s Joe Burns averaged 40.20 from seven matches during the 2017-18 Big Bash League season. Despite playing two less matches than the majority of his team-mates, Burns was the third-highest run-scorer for the Brisbane Heat in 2017-18.

During that season, on January 18, 2018, the Brisbane Heat were bowled out by the Sydney Sixers for an all-time low of 73 in just 16.4 overs. Burns rolled his ankle just before the match and was unable to play. Is it a coincidence that Burns’ pre-match injury cost his side victory? Or is it a telling statistic which proves Burns’ untold value to the Heat?

A batsman’s strike rate is also a crude indication of how efficiently and quickly batsmen score their runs. The traditional batting strike rate provides a total of the expected number of runs a batsman will score after 100 balls. Being able to score runs quickly is an invaluable skill in Twenty20 cricket and therefore, a strike rate of 170 is more desirable than one of 100 because it means more runs will have been scored in the same amount of time.

Like the batting average however, this statistic is not without its flaws. For starters, the batting strike rate spits out the expected number of runs a batsman will score after 100 balls. A game of Twenty20 cricket lasts for 120 balls (20 overs * 6 balls). When was the last time ONE batsman faced 100 balls in a T20 match?

Let’s look at another example. Suppose a batsman came out to bat after the fall of a wicket, hit the first ball he faced for six and then got out with the very next delivery. A quick glance at the strike rate will tell you he was striking at an impressive 300, but a closer analysis of the game scenario will tell you that his team lost 2 wickets for 6 runs off just 3 balls. Therefore, this batsman was actually a detriment to his team for giving up his wicket so easily.

A perhaps more important statistic known as scoring shot percentage is slowly starting to be adapted for use in Twenty20 cricket. Scoring shot percentage tells us the number of balls a batsman scores from as a percentage of total balls the batsman faced. This statistic gives a better idea of a batsman’s ability to turn over the strike than strike rate alone. Players with a high percentage tend to be more versatile players and capable of a wider array of shots. Players of the past such as Simon Katich and Michael Hussey spring to mind.

Both the batting average and strike rate still have their rightful place amongst cricket statistics in the longer formats. In no way are we suggesting that these statistical measurements be thrown out the window. We are simply stating that they should not be used as black and white values to determine a batsman’s true value.

Numbers derived from statistics often need to be put into context. The opposition, a batsman’s home and away record, rate of dot balls and rate of boundaries are other factors that help give these numbers some context and perspective and should also be considered when evaluating batsmen. Luckily for you, that’s where we come in…